Yeah, to me it feels like a step change to a new level, followed by a lot of exploration of that new level. It feels like there's lots of exploring left to do,

Yeah, to me it feels like a step change to a new level, followed by a lot of exploration of that new level. It feels like there's lots of exploring left to do, but we're not going to reach "magic box that turns English into flawless programs" on the current level. Speaking just to programming, yes there's probably lots of AI-generated code in the training data now. But the generation of that code was guided by, and (presumably) confirmed to be working by, a human. So it's been filtered. I suspect that AI-and-human produces better code than human, on average. I know that if I throw any twenty lines of my code into it, it can usually suggest some improvement, even if it's just a variable name.